chore(deps): bump mcp from 1.26.0 to 1.28.1 in /envs/snake_env#979
chore(deps): bump mcp from 1.26.0 to 1.28.1 in /envs/snake_env#979dependabot[bot] wants to merge 1 commit into
Conversation
Bumps [mcp](https://github.com/modelcontextprotocol/python-sdk) from 1.26.0 to 1.28.1. - [Release notes](https://github.com/modelcontextprotocol/python-sdk/releases) - [Changelog](https://github.com/modelcontextprotocol/python-sdk/blob/main/RELEASE.md) - [Commits](modelcontextprotocol/python-sdk@v1.26.0...v1.28.1) --- updated-dependencies: - dependency-name: mcp dependency-version: 1.28.1 dependency-type: indirect ... Signed-off-by: dependabot[bot] <support@github.com>
|
The docs for this PR live here. All of your documentation changes will be reflected on that endpoint. The docs are available until 30 days after the last update. |
There was a problem hiding this comment.
Alignment Review Report
dependabot bump of mcp 1.26.0 → 1.28.1 in /envs/snake_env — lockfile-only (envs/snake_env/uv.lock, +140/-140). mcp is a transitive dependency (openenv → fastmcp → mcp; snake_env pins no mcp), so a lock-only change with no pyproject.toml edit is correct.
Automated Checks
- Lint: PASS for this PR (N/A) —
.claude/hooks/lint.shexits non-zero, but only on ~20 pre-existingenvs/**files unrelated to this change (agent_world_model_env, chat_env, coding_tools_env, jupyter_env, opencode_env, repl_env, terminus_env, textarena_env). This PR touches zero.pyfiles (uv.lock/TOML aren't linted), so it introduces no lint issues.uv lock --check --project envs/snake_envpasses (127 pkgs resolved) — the lock is consistent withpyproject.toml. - Debug code: CLEAN —
check-debug.shhits are all pre-existingsrc/openenv/...prints/TODOs; 0 are in files touched by this PR.
Open RFCs Context
- In Review: 000, 001, 002, 003 (MCP Support — @Darktex/@pankit-eng), 005. Draft: 010. (004 has no status header.)
- RFC 003 is the nearest RFC (this is the MCP Python SDK), but it governs the MCP protocol/interface, not the SDK's pinned version. A transitive 1.26→1.28 bump doesn't conflict. No RFC covers Python packaging / dependency indexes.
Tier 1: Fixes Required
- None. Verified live against PyPI:
mcp1.28.1 hasrequires_python >=3.10(env is>=3.10✓),yanked: false, and 0 known vulnerabilities. Lock hashes match exactly (wheel2726bca5…, sdistd51e36a5…). Package set is unchanged: 125 → 125, 0 added / 0 removed, onlymcpversion changed — no new transitive deps despite the 1.27→1.28 jump.
Tier 2: Alignment Discussion
Principle Conflicts
ALIGNMENT FLAG: Lockfile index source flipped from the HF PyPI mirror to public PyPI (repo-wide in this env)
- Principle/RFC at stake: PRINCIPLES.md — "Container isolation for reproducibility and security" (L15)
- The concern: Beyond the
mcpblock, this re-lock rewrites all 136 packages'source.registryfromhttps://pypi.registries.huggingface.tech/→https://pypi.org/simple(base 136 mirror/0 public → head 0/136), and bumps the lockrevision2→3. Artifact URLs stay onfiles.pythonhosted.organd hashes are unchanged (onlymcp's moved), so it is very likely benign — but it silently changes declared provenance and widens an existing cross-env split (13 envs already on public PyPI, incl. the referenceecho_env; 23 still on the HF mirror). This isn't encoded as policy anywhere (the mirror appears only insideuv.lockfiles — notdependabot.yml, rootpyproject, CI, or docs). Registry choice is a team decision, so flagging rather than fixing. - Suggested reviewer: @Darktex (reproducibility principle author), @burtenshaw (dependabot config owner)
Process note (not a code issue), cc @burtenshaw: .github/dependabot.yml's uv updater is scoped directory: "/" with exclude-paths: ["envs/**"], yet this native dependabot/uv/envs/snake_env/… PR still edits envs/snake_env/uv.lock. exclude-paths is a valid Dependabot key, but the uv ecosystem doesn't appear to honor it here.
Heads-up (latent, won't fire on this PR): revision = 3 needs uv ≥ ~0.8 to parse. PR CI uses uv 0.9.3 (fine), but the openenv-base image ships uv 0.5.27, so a uv sync --frozen rebuild on a rev-3 lock could fail. Lock-only PRs don't trigger docker-build.yml and rev-3 envs already ship, so this is a verify-the-base-image note, not a blocker.
RFC Conflicts
None identified.
Positives
- Security fix:
mcp1.26.0 carries two advisories — GHSA-hvrp-rf83-w775 / CVE-2026-52870 (experimental-tasks session-ownership) and GHSA-jpw9-pfvf-9f58 / CVE-2026-52869 (SSE/Streamable-HTTP session hijack), bothfixed_in: 1.27.2. 1.28.1 remediates both. (OpenEnv's real exposure is low — noenable_tasks()usage — but good hygiene.) - New 1.28.0 deprecations don't affect OpenEnv: 1.28.0 adds
DeprecationWarnings for the MCP-SDK WebSocket transport (mcp.client/server.websocket) and the experimental tasks API. OpenEnv uses none of these (grep ofsrc/+envs/snake_env= 0 hits — and note OpenEnv's WebSocket is its own Gym-orchestration channel, not MCP's transport), and the repo sets nofilterwarnings = ["error"], so no warning escalates to a test failure.
Summary
- 0 mechanical issues to fix
- 1 alignment point for human review (index-source flip / provenance) + 1 process note + 1 latent heads-up
- 0 RFC conflicts
Sent by Cursor Automation: Pre-review
| name = "mcp" | ||
| version = "1.26.0" | ||
| source = { registry = "https://pypi.registries.huggingface.tech/" } | ||
| version = "1.28.1" |
There was a problem hiding this comment.
mcp 1.26.0 → 1.28.1 (transitive via openenv → fastmcp). Verified live on PyPI: requires_python >=3.10, yanked: false, 0 known vulns; lock hashes match exactly (wheel 2726bca5…, sdist d51e36a5…).
This remediates two advisories present in 1.26.0 — GHSA-hvrp-rf83-w775 (CVE-2026-52870) and GHSA-jpw9-pfvf-9f58 (CVE-2026-52869), both fixed_in: 1.27.2.
Note: 1.28.0 newly deprecates the MCP WebSocket transport and the experimental tasks API, but OpenEnv uses neither and sets no filterwarnings=error, so there is no test/CI impact.
| name = "aiofile" | ||
| version = "3.9.0" | ||
| source = { registry = "https://pypi.registries.huggingface.tech/" } | ||
| source = { registry = "https://pypi.org/simple" } |
There was a problem hiding this comment.
Index-source flip — present on all 136 packages in this re-lock: base pinned pypi.registries.huggingface.tech → now pypi.org/simple. Benign (artifacts stay hash-pinned on files.pythonhosted.org; only mcp's hashes changed) but it changes declared provenance repo-wide for this env and widens an existing cross-env split. See the reproducibility flag in the summary. cc @Darktex / @burtenshaw.
| @@ -1,5 +1,5 @@ | |||
| version = 1 | |||
| revision = 2 | |||
| revision = 3 | |||
There was a problem hiding this comment.
Lock revision 2 → 3 (requires uv ≥ ~0.8 to parse). PR CI (uv 0.9.3) and already-shipped rev-3 envs are fine; just a heads-up to confirm the openenv-base image's uv (0.5.27) before a uv sync --frozen rebuild. Not a blocker for this lock-only PR.
|
Aggregated into #921. |
|
OK, I won't notify you again about this release, but will get in touch when a new version is available. If you'd rather skip all updates until the next major or minor version, let me know by commenting If you change your mind, just re-open this PR and I'll resolve any conflicts on it. |


Bumps mcp from 1.26.0 to 1.28.1.
Release notes
Sourced from mcp's releases.
... (truncated)
Commits
777b8d0[v1.x] Support TransportSecuritySettings in the WebSocket server transport (#...4720467[v1.x] Set Development Status classifier to Production/Stable (#2976)6df3d73[v1.x] Buffer per-request StreamableHTTP streams; store priming event before ...32d3290[v1.x] Pass a list to parametrize in test_docs_examples (pytest 9.1.0 compat)...0dca751[v1.x] Deflake the child process cleanup tests (#2839)52258a9[v1.x] Add a v2 status banner to the README (#2835)b8f4917[v1.x] Deprecate the WebSocket transport and the experimental tasks entry poi...2309e5efix: omit null optional fields from task result payloads (#2809)494eb11[v1.x] Support Python 3.14 (#2769)6213787[v1.x] Scope experimental tasks to the session that created them (#2720)Dependabot will resolve any conflicts with this PR as long as you don't alter it yourself. You can also trigger a rebase manually by commenting
@dependabot rebase.Dependabot commands and options
You can trigger Dependabot actions by commenting on this PR:
@dependabot rebasewill rebase this PR@dependabot recreatewill recreate this PR, overwriting any edits that have been made to it@dependabot show <dependency name> ignore conditionswill show all of the ignore conditions of the specified dependency@dependabot ignore this major versionwill close this PR and stop Dependabot creating any more for this major version (unless you reopen the PR or upgrade to it yourself)@dependabot ignore this minor versionwill close this PR and stop Dependabot creating any more for this minor version (unless you reopen the PR or upgrade to it yourself)@dependabot ignore this dependencywill close this PR and stop Dependabot creating any more for this dependency (unless you reopen the PR or upgrade to it yourself)You can disable automated security fix PRs for this repo from the Security Alerts page.
Note
Low Risk
Lockfile-only transitive bump with no snake_env code changes; minor risk if CI treats new mcp DeprecationWarnings as errors when exercising fastmcp/MCP paths.
Overview
Updates
envs/snake_env/uv.lockso the transitivemcpdependency moves from 1.26.0 to 1.28.1 (pulled in viaopenenv→fastmcp). Lock revision increments to 3.The refresh also rewrites package
source.registryentries from the Hugging Face PyPI mirror tohttps://pypi.org/simpleacross the lockfile; resolved versions and wheel hashes otherwise stay aligned with a normaluv lockupdate.No application or
pyproject.tomlchanges—only the lockfile.Reviewed by Cursor Bugbot for commit 1a76930. Bugbot is set up for automated code reviews on this repo. Configure here.